Objective To investigate the impact of different reconstruction algorithms ,including filtered back projection (FBP) , hybrid iterative reconstruction technique (iDose4 ,Philips Healthcare) and iterative model reconstruction (IMR ,Philips Healthcare) on image quality of the low dose chest CT .Methods 30 consecutive patients underwent chest CT on a 256‐slice CT (80 kVp ,80 mAs) were enrolled .Raw data were reconstructed with FBP ,iDose4 ,IMR_soft (L1_Body_Soft_Tissue) and IMR_routine(L1_Body_Routine) algorithms respectively ,and reconstructed thickness was 1 mm ,increment 0 .5 mm .SD ,SNR and CNR were measured and recor‐ded .Image quality was assessed using 4‐point scale (poor to excellent) with lung and mediastinum window ,respectively .We com‐pared quantitative and qualitative parameters among three reconstructions .Results SD of four groups were 55 .7 ± 20 .7 ,37 .0 ± 13 .0 , 13 .4 ± 4 .9 and 19 .5 ± 7 .0 ,repectively .Compared with FBP group ,SD reduced by 33 .5% ,75 .9% ,65% (all ,P<0 .001) in iDose4 , IMR_soft and IMR_routine groups ,respectively .SNR and CNR of IMR_soft and IMR_routine were all higher than FBP group (all , P<0 .001) .SNR and CNR of iDose4 group were higher than FBP group ,but there was no significantly difference (P=0 .08 ,0 .91 , respectively) .SNR and CNR of IMR_soft and IMR_routine groups were all higher than iDose4 group (all ,P<0 .001) ,but was no significantly different between IMR_soft and IMR_routine group(P>0 .05) .The image quality score of lung window in IMR_routine group was the highest in groups ,but was no significantly different with IMR_soft group ,and was significantly different with FBP group and iDose4 group ,respectively .The image quality score of mediastinum window in IMR_soft group was the highest in groups , but was no significantly different with IMR_routine group and was significantly different with FBP group and iDose4 group ,respec‐tively .Conclusion Compared with FBP technique ,iDose4 and IMR techniques can distinctly reduce image noise and improve image quality on low dose CT ,especially IMR algorithm .

Objective To observe the value of iterative model reconstruction(IMR)technique in CT pulmonary angiography (CTPA)with lower radiation dose and lower contrast medium.Methods 60 patients with clinical suspicion of pulmonary embolism underwent 256-slice CTPA were prospectively enrolled.The raw data of test group(n=30)were reconstructed by filtered back projection (FBP)(group A)and IMR technique (group B).And the raw data of control group(n=30)were reconstructed by FBP(group C).The values of volume CT dose index(CTDIvol),effective radiation dose (ED)were recorded and calculated.The image quality parameters:CT value,image noise,signal-to-noise (SNR),contrast-to-noise ratio (CNR)and subjective image score(from one to five)were measured and compared between group A and B,B and C.Results The BMI was no statistically significant difierenee between two groups(P=0.001).The CTDIvol, ED on test group were respectively lower than that on control(P<0.001).The CT value was no statistically significant difierenee between group A and B(P=0.999).The CT valueon group B was higher than that on group C(P=0.005).The image noise on group B was lower than that on group A(P<0.001),but which was higher than that on group C(P<0.096).The SNR,CNR on group B were higher than that on group A(P<0.001).The SNR,CNR did not differ significantly between B and C(P=0.831,P=0.958).The diagnosable rate and good quality rate on group B were higher than that on group A(P<0.001),and there were no significantly diferent between B and C(P=1.000).Conclusion The application of optimized IMR technique can improve image quality at lower radiation dose and lower contrast medium for CTPA.

ObjectiveTo study the feasibility of abdominal aortic CT angiography with low dose contrast medium.MethodsThe patients who took the examinations were divided into three groups.In 34 patients of group A,the injection flow rate was 4 ml/s,and the contrast medium volume was 90 ml and 20 ml of saline was applied.The contrast medium dose and injection flow rate of the other 64 patients were determined according to the formula,injection tlow rate = patients' body mass × ratio of contrast medium dose to the body mass / (5 + exposure time).In B and C group with 32 cases each,the ratio of contrast medium dose to body mass were 0.8 to 0.9 and 0.6 to 0.7 ml/kg.The depiction and the measured intraarterial concentration of contrast medium of the abdominal aorta and the renal arteries as well as their branches were compared and evaluated on the source images and MIP images generated among three groups respectively.CT values of the abdominal aorta and its branches in each group were analyzed using singlefactor analysis of variance,while renal artery and its branches were analyzed using x2 test score.Results The abdominal aorta,the celiac trunk,the splenic artery,the superior mesenteric artery,the renal arteries and their branches were all well shown in three groups.The CT value within the abdominal aorta were (363 ±28),(362 ±28) and (334 ± 14) HU for the three groups respectively,the CT values within the celiac were (368 ±28),(367 ±28) and (338 ± 15)HU,the CT value within the renal artery were (365 ±27),(364 ± 29) and (336 ± 15) HU respectively,the CT value within the splenic artery were (373 ±28),(372 ± 28) and (343 ± 15) HU respectively,and the CT value within the superior mesenteric artery were ( 374 ± 28 ),( 372 ± 28 ) and ( 344 ± 16 ) HU respectively.There were significant differences among the three groups ( P = 0.000).There were significant differences between A group and C group,B group and C group ( P ＜ 0.01 ),while no significant difference existed between A group and B group( P ＞0.05 ).The amount of contrast agent for each patient in group A was 90 ml,while the amount of contrast agent in group B and C was (51 ± 9) and (40 ± 9) ml respectively.There were significant differences of the contrast medium volume among the three groups ( F = 537.091,P = 0.000).Results of pairwise comparisons were statistically significant(P =0.000).Image quality of renal artery in group A was rated as excellent,good,and generally for 26,7 and 1 cases,24,7 and 1 for group B,and 22,8 and 2 for group C.There were no significant differences in image quality of the renal arteries among the three groups ( x2 = 0.561,P = 0.755 ).ConclusionThe individualization of injection method in abdominal aortic CT angiography can minimize the amount of contrast agent but to meet the clinical requirements.

Objective To investigate the feasibility of contrast-enhanced CT with low tube voltage using iterative model reconstruction (IMR) technique.Methods Sixty patients were randomly assigned into 2 groups (group A and group B, 30 each) according to random number table.All patients underwent contrast-enhanced hepatic CT.Group A was scanned with 100 kV at arterial phase(AP) and 120 kV at portal vein phase (PVP), while group B was scanned with 120 kV at AP and 100 kV at PVP.All protocols were performed at the same tube current of 250 mAs.Raw data were reconstructed with IMR for AP images in group A and PVP images in group B;and reconstructed with FBP for AP images in group B and PVP images in group A.Images of 4 different groups were obtained: A1(AP,100 kV,IMR) , B1(AP,120 kV, FBP), A2(PVP, 120 kV,FBP) and B2(PVP, 100 kV, IMR).Subjective evaluation indexes for image quality including low-contrast detectability, lesion edge sharpness, image distortion and diagnostic confidence.Objective evaluation indexes included CT attenuation of hepatic parenchyma, image noise, SNR and CNR, which were assessed and compared between groups A1 and B1, groups A2 and B2.Effective radiation doses were calculated.Results Effective dose in group A1 was reduced 35.1％ compared toB1 (t=ll.05, P＜0.01), while a reduction of 37.7％ in group B2 compared to A2 (t=11.64,P ＜ 0.01).Subjective image quality score of low-contrast detectability and lesion edge sharpness were significantly higher in group A1 compared to B1 (Z =6.391, 3.200, P ＜ 0.01), as well as in group B2 compared to A2 (Z =6.559, 3.409, P ＜ 0.01).No differences were found in image distortion and diagnostic confidence between groups A1 and B1, groups A2 and B2, respectively (P ＞ 0.05).Significantly lower image noise and higher SNR/CNR were found in group A1 compared to group B1 (t =12.889, 15.458, 1.325, P ＜ 0.01) , as well as in group B2 compared to group A2(t =15.163, 15.308, 3.136, P ＜0.01).Conclusions Significant radiation dose reduction and image quality improvement in contrast-enhanced hepatic CT can be reached by using low tube voltage protocol combining with IMR technique.